Using the Solution Space Diagram in Measuring the Effect of Sector Complexity During Merging Scenarios
AIAA Guidance, Navigation, and Control Conference
When designing Air Traffic Control (ATC) sectors and procedures, traffic complexity and workload are important issues. For predicting ATC workload, metrics based on the Solution Space Diagram (SSD) have been proposed. This paper studies the effect of sector design on workload and SSD metrics. When considering the SSD in evaluation of a sector, each aircraft within the sector introduces a zone of conflict, the Forbidden Beam Zone (FBZ), on the SSD. The properties of these FBZ are systematically
... are systematically studied to increase understanding of the SSD usability in assessing workload and sector complexity. The effects of sector design variables on Air Traffic Controller (ATCo) workload and also SSD properties were evaluated. Example of sector properties are, number of streams to be merged, the merge angle, the proximity of incoming aircraft and the variability of traffic mix of small and large aircraft. Based on the findings, each sector design variable leads to different effect on both workload and SSD properties. Apart from that, correlation between the workload and the SSD properties were found to be in a higher level than of the number of aircraft within the sector, proving that the SSD-based analysis to be a good workload indicator. These correlations were studied based on two different groups of subjects with ranging experience in order to demonstrate the robustness of the method. 2 through primary and secondary task performance, subjective workload assessment using continuous and discontinuous workload ratings and also using physiological measures of the mental workload. In order to understand workload in relation to sector design a method to quantify ATCo's workload is needed. One example is by using the sector complexity as an objective measurement indicator. Aircraft count or aircraft density is one method of determining sector complexity by looking at the number of aircraft within a sector. It is defined as the number of aircraft per unit of sector volume. Experiments showed that of all the individual sector characteristics, aircraft density shows the highest correlation with ATCo subjective workload ratings. 6,7 However, aircraft density has significant shortcomings in its ability to accurately measure and predict sector level complexity 6,8 where it is unable to capture the dynamic behavior of aircraft in the sector. For example, five aircraft flying in the same direction do not exhibit the same complexity as the same number of aircraft flying in various directions. Another measurement of sector complexity is dynamic density, which incorporates the dynamic behavior of aircraft in the sector. Research on dynamic density by Laudeman et al. 9 and Sridhar et al. 10 has indicated a number of variables for calculating the dynamic density and each factor is given a subjective weight. The dynamic density is a summation of these variables and its corresponding subjective weight. However, the summation of dynamic density that is based on subjective weights gathered from regression methods on samples of traffic data means that the method only be used on scenarios that do not deviate too much from the baseline scenario. Therefore the dynamic density metric is not generally applicable to just any situation. 11 In order to better assess airspace complexity, new methods such as 'complexity maps' and the 'solution space' are proposed by Lee et al. 12 and Hermes et al. 11 , respectively. Both solutions act as an airspace complexity measure method, where a complexity map details the operator's control activity as a function of the parameters describing the disturbances, and the solution space details those two-dimensional speed and heading possibilities of one controlled aircraft that will not induce separation violation. A. Sector Complexity and Workload The sector complexity construct is described based on the important factors associated with a sector's structural and flow characteristics. The structural characteristics are fixed for a sector such as terrain, number of airways, airway crossings and navigation aids. The flow characteristics, however, vary as a function of time and depend on characteristics like number of aircraft, mix of aircraft, weather, separations between aircraft, aircraft speeds and flow restrictions. A combination of these structural and flow parameters influences the airspace demand, thus the controller's workload. 10 Figure 1 shows the relationship between taskload and workload as described by Hilburn and Jorna 4 , where we adapted the position of sector complexity within the diagram. The function of the Solution Space Diagram (SSD) is included as a workload measure and alleviator, possibly also affecting the sector planning. Given that sector complexity having relation with the taskload and thus workload as experienced by the controller, a sound research has to be performed in order to have a better understanding of the sector complexity measure and its effect to workload, performance and safety. Majumdar and Ochieng 13 , in their research has defined a total of over 50 complexity-shaping factors (CSF). Having said that there are numerous sector complexity variables that can influence controller's workload, a good sector design has to be achieved. In order accomplish that, the impact of the Air Traffic Control (ATC) complexity variable on the controller's workload has to be assessed systematically. Several methods and metrics have been proposed to quantify the complexity of a sector, and among all is the SSD. The SSD has been shown to be useful as a good measure for workload, and the presentation of the diagram to ATCos even alleviated them in previous researches by D , respectively. In this paper, the investigations of sector design variables were included. Figure 1: Taskload and Workload (adapted from Hilburn and Jorna, 2001). Downloaded by TECHNISCHE UNIVERSITEIT DELFT on February 28, 2013 | http://arc.aiaa.org | The SSD is used as a basis for obtaining a better metric for complexity, capable of predicting the task demand load of air traffic controllers. In previous research, the solution space-based metric proved to be a more objective and scenario-independent metric than a weighted combination of scenario properties, in which the weights are highly dependent on the baseline scenarios considered. 11, 14, 15 Initial work by Van Dam et al. 17 has introduced the application of Solution Space in aircraft separation problems from the pilot's perspective. Hermes et al. 11 , have continued the idea of using solution space in aircraft separation problem but for the Air Traffic Control (ATC) problem. Based on research conducted by Hermes et al. 11 and D'Engelbronner et al. 14 , a high correlation was shown to exist between the Solution Space and ATCo's workload. study the workload from a different perspective, looking at possibility of using the SSD as an interface to reduce controller's workload. Based on his studies, 15 he indicates that the diagram could indeed reduce controller's workload in a situation of increased traffic level. Results also show the possibility of the existence of a traffic threshold level, up to which the SSD interface is effective in reducing workload. This paper follows the proposed method by Hermes et al. 11 , and adds an exploration of the effect of sector design variables on the controller's workload and the metrics based on the SSD. To this end, an experiment was conducted in which the traffic pattern was consistently varied. The SSD was used as an offline evaluation method of sector complexity and workload, and metrics based on the SSD were compared to workload ratings given in the experiment. Other performance related measurement such as the number of commands, number of conflicts, extra distance ratio, smallest aircraft separation et cetera are also evaluated in order to see the correlation between the performances of the subjects to the workload indicated. A questionnaire regarding the controller perceived workload and also comments regarding the experiment were also taken at the end of every simulation run. Based on the initial quantitative study conducted, it is concluded that different sector variable gives different workload pattern. 16 In this experiment, it is expected that the differences between expert and non-expert subjects' behavior are uncovered and most importantly the workload behavior towards sector complexity variable is identified.